What Do Titano- and Zirconocenes Do with Diynes and Polyynes

Acc. Chem. Res. 2000, 33, 119-129

What Do Titano- and Zirconocenes Do with Diynes and Polyynes?‡ UWE ROSENTHAL,* PAUL-MICHAEL PELLNY, FRANK G. KIRCHBAUER, AND VLADIMIR V. BURLAKOV§ Institut fu ¨ r Organische Katalyseforschung, an der Universita¨t Rostock, Buchbinderstrasse 5-6, D-18055 Rostock, Germany Received June 11, 1999 ABSTRACT A thorough investigation of the reactions of diynes, R(CtC)2R, and polyynes, R(CtC)nR, with titanocene and zirconocene was performed using the metallocene sources Cp2M(L)(η2-Me3SiCtCSiMe3) (M ) Ti, L ) -; M ) Zr, L ) THF, pyridine). The conversions show an array of different products generated in complexation, coupling, and cleavage reactions. These results include remarkable structures (e.g., five-membered metallacyclocumulenes) and reactions (e.g., C-C single-bond cleavage). In addition, the first C-C single-bond metathesis in homogeneous solution was discovered. The presented findings have been supported by theoretical studies.

Introduction A suitable complex fragment that is both coordinatively and electronically unsaturated is required to realize a variety of stoichiometric and catalytic reactions of organometallic compounds with different substrates. Such a

Uwe Rosenthal, born in 1950 in the Altmark, East Germany, received his Ph.D. (1976) under the supervision of E. Kurras and his habilitation (1989) at the University of Rostock. In 1988 he pursued postdoctoral work at the A. N. Nesmeyanov Institute of Organoelement Compounds of the Russian Academy of Sciences in Moscow with M. E. Vol’pin and V. B. Shur. After the Berlin wall was torn down he became a visiting research scientist at the Max-Planck-Institut fu¨ r Kohlenforschung in Mu¨ lheim/Ruhr, working a full year with G. Wilke and K. R. Po¨ rschke (1990/91). From 1992 to 1996 he headed the Max-Planck Research Group “Complex Catalysis”, and in 1993 he became Professor of Inorganic Chemistry at the University of Rostock, working at the Institut fu¨ r Organische Katalyseforschung. While he is broadly interested in organometallic chemistry and complex catalysis, his research interests focus on the activation of alkynes, olefins, and small molecules, as well as of C-C and Si-H bonds. He has published 115 papers in scientific journals. Paul-Michael Pellny, born in Du¨ sseldorf, Germany, in 1968, studied chemistry at the Technical University of Darmstadt (1989-96) and obtained his M.Sc. degree (Dipl.-Ing.) with J. Gaube (1996). From 1997 until November 1999 he worked on his Ph.D. thesis with U. Rosenthal at the Institut fu¨ r Organische Katalyseforschung, concentrating on the activation of di- and polyynes. Frank G. Kirchbauer was born in Schwenningen, Germany, in 1971. He studied chemistry at the University of Konstanz (1990-96) and obtained his M.Sc. (Dipl.Chem.) with H. Fischer (1996). In 1997 he joined the Rosenthal group at the Institut fu¨ r Organische Katalyseforschung to start working on his Ph.D. thesis, which he concluded very recently. His research interests include C-C single-bond activation and metathesis. Vladimir V. Burlakov was born in Moscow, Russia, in 1955 and studied chemistry at the Moscow Academy of Fine Technology (1976-82), receiving his M.Sc. degree with O. N. Temkin (1982). After moving to the A. N. Nesmeyanov Institute of Organoelement Compounds of the Russian Academy of Sciences in Moscow and finishing his Ph.D. with V. B. Shur (1995), he became a senior researcher in the same institute. His interests focus on the transformations of unsaturated hydrocarbons and small molecules mediated by titanocene and zirconocene complexes. He has published 85 papers in scientific journals and has been, for altogether more than three years during the last eight years, a visiting research scientist in the Rosenthal group (1991-98). 10.1021/ar9900109 CCC: $19.00 Published on Web 01/25/2000

 2000 American Chemical Society

core complex coordinates to the chosen substrates, activates them, and moves them in the direction of the desired products. This general mechanism is nicely illustrated in manifold reactions of group 4 metallocenes. The complex fragments titanocene, “Cp2Ti”, and zirconocene, “Cp2Zr”, are unstable 14-electron compounds displaying a d2 configuration (M(II)). They possess one lone-electron pair and two vacant valence orbitals, making their reactivity comparable to that of carbenes. The possible interactions between occupied and unoccupied orbitals explain why these metallocenes, “Cp2M”, react with a variety of unsaturated compounds to form metallacycles which then have the potential to undergo diverse conversions with further substrates. An important question in this context is what kind of ligand can be used that sufficiently stabilizes the metallocene fragment and that can be released quantitatively under mild conditions to generate the unstable and very reactive core complex. Possible answers to that problem were provided not only by an array of combined systems of Cp2MX2/reducing agent but also by some well-defined complexes Cp2MLn (e.g., M ) Zr, L ) n-butene, n ) 1; M ) Zr, L ) 1,3-butadiene, n ) 1; M ) Ti, L ) PMe3, n ) 2; etc.). These complexes are discussed in a number of reviews and various contributions in textbooks and are appreciated with regard to their importance in synthetic chemistry.1 The applicability of these systems and their success in certain reactions often depend on their preparative accessibility, on the selectivity of the conversions, and on the inertness of the stand-in ligand. In recent years, we have successfully established alkyne complexes of the type Cp2M(L)(η2-Me3SiC2SiMe3) (M ) Ti, L ) -; M ) Zr, L ) THF, pyridine) as novel reagents for group 4 metallocene fragments. These react with a multitude of substrates, and in many cases the products explicitly differ from those obtained with conventional metallocene sources.2 This can best be explained on the basis of the ready release of the silyl-substituted alkyne. For M ) Zr, strong evidence has been found for an associative mechanism,2c while for M ) Ti, the substitution can also run dissociatively via a free titanocene. The first examples of free titanocenes have very recently been published.3 Group 4 metallocenes play an important role in the stereospecific polymerization of olefins. For that reason, olefins have been the substrates of choice, and these were mainly studied in reactions with the metallocenes. In the range of a “modern acetylene chemistry”,4 which serves a variety of implementations (e.g., new materials, special polymers, nanostructures, supramolecular structures, etc.), substrates with C-C triple bonds have increasingly been investigated. After originally focusing on monoalkynes, researchers started to also concentrate on di-, oligo-, and polyynes. These substrates cause a greater complexity and ‡ This Account is dedicated to Professor Gu ¨nther Wilke on the occasion of his 75th birthday. * To whom correspondence should be addressed. Phone/fax: 0049381-4669376. E-mail: [email protected]. § On leave from the Institute of Organoelement Compounds of the Russian Academy of Sciences, Moscow, Russia.

VOL. 33, NO. 2, 2000 / ACCOUNTS OF CHEMICAL RESEARCH

119

Titano- and Zirconocenes with Diynes and Polyynes Rosenthal et al.

Scheme 1

a decreased transparence of the reactions. But this drawback is compensated by the considerably greater attractiveness of the targeted structures which is due to the potential of spatial cross-linking. The literature covers a large number of examples dealing with coupling and cyclization reactions involving low-valent titanium and zirconium compounds.1 These reactions and those of diynes with other unsaturated substrates will not be covered in this paper. This Account exclusively considers products of the reactions of di-, oligo-, and polyynes with titano- and zirconocene. Emphasis is put on the conversions of 1,3-butadiynes with metallocenes which show remarkable, unusual structures (e.g., five-membered metallacyclocumulenes) and reactions (C-C single-bond cleavage and metathesis), while a decisive influence is exerted by the metals used, the diyne substituents, and the stoichiometry. These results illustrate again that the combination of transition metal compounds, substrates, and stoichiometry can lead to a fascinating and novel chemistry. The aim of this article is to give an overview of this exciting area of chemistry. So, what happens when titanocene and zirconocene react with these unsaturated moieties?

Nonconjugated Diynes and Triynes The diversity of reactions of nonconjugated diynes RCt CsXsCtCR with titanocene and zirconocene is very

Scheme 2

120 ACCOUNTS OF CHEMICAL RESEARCH / VOL. 33, NO. 2, 2000

Scheme 3

extensive and therefore will be covered here just briefly. Principally, a differentiation between two possible reaction pathways can be made: we find coupling and complexation reactions. In the case of reductive cyclizations, the exo-exo coupling to afford metallacyclopentadienes is dominant (Scheme 1). The coupling to mononuclear complexes has been found with a great variety of different bridges X1,5 (e.g., M ) Zr, X ) (CH2)2, (CH2)4,1b M ) Zr, X ) CH2OCH2,5a PR,5b B(NR2)-CH2-B(NR2)5c). Besides that, polymers, cyclotrimers, cyclotetramers, and macrocycles containing metallacyclopentadiene units could be synthesized, dependent on X6 (e.g., M ) Zr, X ) p-Me2SiC6H4SiMe2,6a Me2SiC6H4-C6H4SiMe2,6a C6H4C6H4,6b bipyridyl6c). Rather strikingly, the coupling for M ) Zr and X ) CH2 leads to binuclear zirconacyclopentadiene complexes7 (Scheme 2). Bis(alkynyl)silanes (X ) SiR2) afford annelated zirconacyclobutene-silacyclobutene compounds with Zr (Scheme 2),8 whose formation can be rationalized by an initial coupling to zirconacyclopentadienes and a subsequent rearrangement. A 1:1 complexation results from R ) SiMe3 and X ) o-C6H4. A coordination of the triple bond is observed for M ) Zr, which does not lead to coupling (Scheme 2).9 When organometallic diynes, e.g., X ) Cp2Zr, are used, the reaction with zirconocene affords σ,πacetylides (Scheme 3).1h,27c The nature of X seems to exert the decisive influence on which product is formed in these reactions. The titanocene reagent Cp2Ti(η2-Me3SiC2SiMe3) (Figure 1) also forms mononuclear titanacyclopentadienes when it is reacted with nonconjugated diynes (X ) (CH2)n, n ) 2-6). In this context, it is of interest that the respective compound with n ) 4 rearranges, via a Cp cleavage and


Scheme 5

FIGURE 1. Molecular structure of Cp2Ti(η2-Me3SiC2SiMe3). an intramolecular C-C coupling, to a stable η4,η3-dihydroindenyltitanium complex.10 The structural details of this compound reveal a strong (C-C)-Ti agostic, as seen by a comparison to very similar complexes (Scheme 4).10,11 The coupling of the nonconjugated triynes RSi(C6H4CtCSiMe3)3 at a zirconocene center lead, in a simple and effective way, to well-defined cage compounds.12

Reactions of 1,3-Butadiynes with Titanocene and Zirconocene An overview of the products detected and fully characterized by the authors is presented in Scheme 5. In the following, details concerning the modes of formation of these complexes will be discussed. Coupling Reactions. The reaction of bis(trimethylsilyl)butadiyne with zirconocene leads to a coupling product of two diyne entities at the core complex. The description of the formed product as a seven-membered zirconacyclocumulene (zirconacyclohepta-2,4,5,6-tetraene) follows from its determined X-ray structure (Scheme 6).13 Using analogous conditions with titanocene and employing 2 equiv of the diyne, a regioselective coupling of the diynes to a titanacyclopentadiene is observed, the product exhibiting one alkynyl group in the R- and another one in the β-position to the metal (Scheme 6).14 It seems rather plausible that a five-membered ring is also formed for zirconium but that this zirconacyclopentadiyne then rearranges to the zirconacycloheptatetraene. This additional reaction step involving the alkynyl group in the R-position is obviously not feasible for the smaller titanium.

In addition to the described coupling of two diynes at a single metal center, there are examples of coupling reactions of two diynes between two metallocene fragments. These will be covered later in this text. C-C Single-Bond Cleavage. Only changing the stoichiometry of the conversion can alter the reaction path-

Scheme 6

Scheme 4

VOL. 33, NO. 2, 2000 / ACCOUNTS OF CHEMICAL RESEARCH 121


Scheme 7

way, and the above-mentioned coupling reactions can be transformed to cleavage reactions if 2 equiv of the metallocene reacts with certain diynes (Scheme 7, upper part).15 The generated products are, from a formal point of view, twofold σ,π-alkynyl-bridged metal(III) complexes. These were previously known and thoroughly investigated, although they were always synthesized differently. A C-C single-bond activation in connection with a cleavage of that bond was a novel way to realize these compounds starting from 1,3-butadiynes. This type of cleavage reaction is favored for M ) Zr15b as compared to Ti,15a an observation that is backed by theoretical calculations.16 Additionally, there is a drastic influence of the substituents R attached to the butadiynes RCtCsCtCR, as exemplified by the trimethylsilyl group, which profoundly activates the inner C-C single bond by its β effect (withdrawing electron density at the C-β atoms). This explains why, for M ) Ti, the bis(trimethylsilyl)butadiyne is the only diyne substrate which could have been cleaved so far. Different Modes of Complexation. For M ) Ti and diverse butadiynes, binuclear complexes with intact C4 units between the two metal centers are formed under analogous conditions. The former diynes are transformed to “zigzag butadiyne ligands” or µ-η(1-3),η(2-4)trans,trans-tetradehydrobutadiene moieties between two metallocene cores (Scheme 7).17 The first representative of this type of complex was obtained by Teuben and De Liefde Meijer in the reaction of [Cp2TiCl]2 with PhCtCNa, which originally was assumed to be [Cp2TiCtCPh]2.17a Later this complex was identified by Sekutowsky and Stucky as a dinuclear complex with a bridging butadiyne, as shown in the lower part of Scheme 7.17b So far, this

Scheme 8

122 ACCOUNTS OF CHEMICAL RESEARCH / VOL. 33, NO. 2, 2000

Scheme 9

bond type is not known for M ) Zr, although theoretical calculations predict its existence in the case of certain substituents (e.g., CN).18 Besides this 2:1 complexation of the titanocene, a 1:1 complexation is observed when a different stoichiometry is employed. The butadiynes RCtCsCtCR with R ) tBu14 and Ph19 generate the novel five-membered titanacyclocumulenes (metallacyclopenta-2,3,4-trienes, η4-1,2,3-butadiyne complexes) (Scheme 8). For R ) tBu, this structural type could also be realized with M ) Zr.20 Scheme 10


Scheme 11

Structural and Theoretical Description of the Metallacyclocumulenes. A five-membered metallacyclocumulene is very unusual. The structures of this type of compound display an almost planar arrangement of the metallacycle, containing three C-C double bonds, of which the central one is elongated. This elongation is ascribed to the intramolecular interaction of this bond to the metal center (Scheme 9).14,20 The distances and the angles (70-74° at C-R and 147150° at C-β) in the metallacycle correspond closely to those of calculated organic cyclocumulenes such as cyclohexa- and cycloheptatrienes.21 According to other theoretical calculations,22 titana- and zirconacyclocumulenes are thermodynamically more stable than the isomeric bis(σ-acetylide) complexes. The calculated data are in good agreement with the obtained experimental values. All four carbon atoms of the former diyne are viewed to have p orbitals perpendicular to the plane of the cyclocumulene. The sp-hybridized internal C atoms possess additional p orbitals in that plane, which are used to establish a coordination of that bond to the metal center (Scheme 9). Stoichiometric Reactions of the Metallacyclocumulenes. Regarding some reactions of the metallacyclocumulenes, it is possible to assume an equilibrium between a η4-complex (metallacyclocumulene) and a η2-complex (metallacyclopropene).19 For instance, the phenyl-substituted titanacyclocumulene reacts with acetone and water to yield insertion products that are typical for titanacyclopropenes (Scheme 10).19 Evidently, both forms of the equilibrium mixture react even with each other to afford an unsymmetric complex,

Scheme 12

Scheme 13

where a titanacyclopentadiene is annelated to a titanacyclopentene (Scheme 11).19 This can be rationalized by an insertion of the internal double bond of the titanacyclocumulene into the titanacyclopropene. Additionally, the symmetric titanium-substituted radialene is generated in the same solution. This can be thought of as a formal dimerization of two titanacyclocumulene molecules (Scheme 11).18,23 If the titanacyclocumulene is treated with zirconocene or, the other way around, the zirconacyclocumulene with titanocene, identical Ti-Zr bimetallic complexes with two σ,π-alkynyl bridges are afforded (Scheme 12).14 The conversions with nickel(0) complexes prove a pointed dependence on the substituents R. For M ) Ti and Zr and R ) Ph, π-complexes of the corresponding metallacyclocumulenes are formed. They can be regarded as Ni(0)-olefin complexes (Scheme 13)24 and are significant for two reasons. On one hand, the homobimetallic analogues of these compounds can be viewed as the intermediates for the consecutive reaction to form the two different 2:1 complexes (cleavage and complexation reaction) (Scheme 7). On the other hand, they are a real complex-chemical proof for the metallacyclocumulene structure of the η4-complex.24,25 Meanwhile, such π-complexes of other metals with different metallacyclocumulenes are known.26 Scheme 14



Scheme 15

Photochemical Investigations. When a solution of the complex Cp2Ti(σ-CtCtBu)2 is irradiated, a titanacyclocumulene is, as shown by NMR spectroscopy, generated in the first step. Consequently, this species reacts with a further equivalent of titanocene to yield the dinuclear complex with µ-η(1-3),η(2-4)-trans,trans-butadiene units between two metallocene cores (Scheme 14).24 This reaction cannot be used preparatively to synthesize the metallacyclocumulenes, as this intermediate is nonisolable. It remains unclear whether in the first step of the reaction the complex [Cp2Ti(σ-CtCtBu)] or the fragment [Cp2Ti] and the radicals [•CtCtBu] are formed, as both possibilities could account for the observed reaction course. This reaction was postulated some time ago27a and was later found to also occur under the catalytic influence of certain Lewis acids.27b Generalized Reaction Scheme. By taking into account the results of our investigations24 as well as excellent theoretical calculations,22 a general simplified scheme is deduced (Scheme 15).24 According to this scheme, the cleavage as well as the coupling progresses via metallacyclocumulenes, in which the intramolecular coordination of the inner double bond is, in the subsequent step, replaced by an intermolecular one. These intermediates then rearrange, depending on the metals M and the substituents R, to afford products in which either an intact or a cleaved C4 linkage is present. The individual energy levels of these and comparable complexes as well as of the supposed intermediates have been determined by thorough calculations, and the results confirm the relative thermodynamic stabilities as observed experimentally. Thus, these theoretical results emphatically support the documented conceptions about the reaction course.22 124 ACCOUNTS OF CHEMICAL RESEARCH / VOL. 33, NO. 2, 2000

Photocatalytic C-C Single-Bond Metathesis. A cleavage of various symmetrically substituted butadiynes in combination with a subsequent alternating recombination of the fragments realizes a C-C single-bond metathesis.28 It is of special note in this context that the first C-C singlebond metathesis reaction ever published has very recently been accomplished using paraffins and immobilized Ta-H catalysts.29 If one-to-one mixtures of the butadiynes tBuCtC-CtCtBu and Me SiCtC-CtCSiMe are treated 3 3 with an excess of the Cp2Ti reagent and that mixture is then irradiated, the unsymmetrically substituted diyne tBuCtCsCtCSiMe is afforded after an oxidative workup, 3 in addition to the symmetrically substituted starting diynes (Scheme 16). The observed reaction does not work without titanocene. If no irradiation is applied, higher temperatures are essential, and a considerable increase in decomposition reactions is observed. This makes this reaction the first titanocene-mediated, photocatalyzed C-C singlebond metathesis in homogeneous solution.28 With regard to the metal compound, this metathesis cannot be conducted catalytically, because an excess of the diyne would favor coupling reactions, making those complexes dominant in the product range (see above). The course of the reaction (Scheme 17) can be formulated in such a way that the titanocene reacts with tBuCtCsCtCtBu to form the binuclear complex with an intact C4 backbone and

Scheme 16


Scheme 17

with Me3SiCtCsCtCSiMe3 to form the σ,π-alkynylbridged cleavage product. Under the influence of light, both 2:1 complexes are subsequently cleaved to the extremely unstable monomeric Ti(III) complexes [Cp2Ti(σ-CtCtBu)] und [Cp2Ti(σCtCSiMe3)], which then dimerize to either the respective starting complexes or, in the desired way, to a differently substituted binuclear complex. The reverse reaction can be realized as well. But there remains the drawback of the noncatalytic reaction sequence with respect to the metal compound. As is well established, the use of sterically demanding ligands can often restrict, if not totally avoid, unintentional coupling reactions of the substrates. We therefore conducted experiments, in which the pentamethylcyclopentadienyl ligand (Cp*) was substituted for the cyclopentadienyl ligand (Cp). Permethylzirconocene and 1,3-Butadiynes. All attempts to realize a catalytic C-C single-bond metathesis of butadiynes using the permethylmetallocene complexes Cp*2M(η2-Me3SiCtCSiMe3) (M ) Ti, Zr) proved unsuccessful, and no stoichiometric products could be recovered in these experimental studies. The unstable monomeric Ti(III) complexes [Cp2Ti(σ-CtCR)] seem to play a key role in the reaction sequence of our metathesis reaction. The corresponding permethyltitanocene complexes Cp*2M(σ-CtCR) (M ) Ti, Zr; R ) Ph, tBu, SiMe3) could not be generated via a salt elimination reaction starting from Cp*2TiX and LiCtCR. For that reason the stable M(IV) complexes Cp*2M(σ-CtCR)2 (M ) Ti, Zr; R ) Ph, tBu, SiMe3) were synthesized, and a reduction, which is well-known for the analogous Cp complexes, should have led to the targeted compounds. During the performance of these investigations, it became evident that, for

M ) Zr, these complexes very easily rearranged in plain sunlight to afford the zirconacyclocumulenes (Scheme 18).30 Additionally, we were able to establish an alternative synthetic route to some novel zirconacyclocumulenes: the reduction of Cp*2ZrCl2 with magnesium in the presence of the butadiynes RCtCsCtCR (R ) Me, Ph, SiMe3) is suitable to generate the η4-complexes (Figure 2), while this does not work for R ) tBu (compare to Scheme 20). Permethyltitanocene and 1,3-Butadiynes. Both of the described methods fail to prepare the corresponding titanacyclocumulenes. For R ) SiMe3, a titanacyclopropene, a η2-complex, is formed instead, probably due to steric reasons (Scheme 19).31 Manifold C-H activations of the methyl groups of the pentamethylcyclopentadienyl ligand and C-C couplings with the particular diynes have been verified for M ) Zr, R ) tBu; M ) Ti, R ) tBu; and M ) Ti, R ) Me, Ph (Scheme 20). These reactions are clearly dependent on the metals M and the substituents R; they are, however, worthless in regard to the synthesis of metallacyclocumulenes and to their application in the catalytic C-C single-bond metathesis. Very interesting, though, are the consecutive reactions of the permethylmetallocene complexes of the butadiynes with carbon dioxide (Scheme 21). While the zirconacyclocumulene (η4-complex) takes up

Scheme 18

FIGURE 2. Molecular structure of Cp*2Zr(η4-1,2,3,4-Me3SiC4SiMe3). VOL. 33, NO. 2, 2000 / ACCOUNTS OF CHEMICAL RESEARCH 125


Scheme 19

Scheme 20

two molecules of CO2 and thereby forms a cumulenic dicarboxylate (Figure 3), the titanacyclopropene (η2complex) inserts only one molecule of CO2 under the formation of a titanafuranone (Scheme 21). It remains unclear whether the zirconacyclocumulene reacts as a η4- or a η2-complex. The reaction course

illustrated in Scheme 21 seems possible. According to that proposal, an analogous zirconafuranone can be formed starting from a η2-complex, and only for zirconium a further reaction of the alkynyl group in the R-position to the metal is possible. The insertion of a second molecule of CO2 can also be prevented by employing a sterically very demanding substituent on the diyne (e.g., tBuMe2Si) (Scheme 22). Altogether, the stability and reactivity of five-membered metallacyclocumulenes appear to be profoundly influenced by steric factors. These can originate from the steric

Scheme 21

FIGURE 3. Molecular structure of Cp*2Zr[sOC(dO)C(SiMe3)dCd CdC(SiMe3)C(dO)Os]. 126 ACCOUNTS OF CHEMICAL RESEARCH / VOL. 33, NO. 2, 2000


Scheme 22

Scheme 24

Scheme 23

demand of the Cp and Cp* ligands, respectively, and the size of the metal as well as of the substituents attached to the diyne. Linear and Branched Polyynes. The experiences with the reactions of the alkynes RCtCR and 1,3-butadiynes R(CtC)2R with titanocenes und zirconocenes can be used to apply the observed reactivity pattern to oligo- und polyynes, R(CtC)nR. In doing this, we aimed at modeling the interactions of group 4 metallocenes with the simplest carbon modification, the “carbyne” (-CtC-)n. Principally, one has to differentiate between two reaction pathways: the complexation and the cleavage reactions. In the

conversion of bis(trimethylsilyl)octatetrayne with 2 equiv of titanocene, a complexation is observed, which is shown in Scheme 23.32 If 4 equiv of the metallocenes are reacted with the octatetraynes RCtCsCtCsCtCsCtCR, the products obtained decisively depend on the substituents R and the metals M (Scheme 24).

Scheme 25



Using R ) SiMe3 with the tetrayne, both titanocene and zirconocene lead to complexes in which the two outer C-C single bonds are cleaved (Scheme 24). Changing to R ) tBu and applying titanocene, we were able to prevent a cleavage and maintained an intact C8 backbone in the formed 4:1 complex (Scheme 24). The found reactivity pattern can also be applied to branched polyynes such as the 1,3,5-tris(tert-butylbutadiynyl)benzene (tBuCtCsCtC)3C6H3 (Scheme 25).33 In the conversions with the metallocenes, we observe a coordination of six titanocene cores to an intact polyyne in the case of M ) Ti, while for M ) Zr altogether three C-C single bonds are cleaved in the 6:1 complex. These and related compounds seem to have potential as materials for nonlinear optical devices, in multicenter catalysis, and as precursors for dendrimers.

Conclusions The metallocene sources Cp2M(L)(η2-Me3SiCtCSiMe3) (M ) Ti, L ) -; M ) Zr, L ) THF, pyridine) are very suitable and convenient reagents for the investigation of reactions of diynes R(CtC)2R and polyynes R(CtC)nR with titanocene and zirconocene. The exceedingly multifaceted chemistry is epitomized by a variety of complexation, coupling, and cleavage reactions of the substrates employed. The course of the reaction can effectively be influenced by varying the metals M, the substituents R, and the stoichiometry. The inquiries into the conversions of simple butadiynes served as a model, and this led to the establishment of the first photocatalyzed C-C singlebond metathesis in homogeneous solution and, additionally, to the application of the observed reactivity pattern to more extensive substrates such as linear and branched oligo- and polyynes. The reactions of the butadiynes have generated some particularly surprising and uncommon structures, of which the metallacyclocumulenes occupy a key role. Even some extensive and profound theoretical studies in that field were initiated by this experimental work. It has repeatedly been proven that the combination of organometallic complexes with suitable substrates very often leads to novel bond types and reactivities. This work was supported by the Max-Planck-Gesellschaft, Deutsche Forschungsgemeinschaft, Fonds der Chemischen Industrie, and the federal state of Mecklenburg-West Pomerania. Funding and facilities provided by the Institut fu ¨ r Organische Katalyseforschung at the University of Rostock are gratefully acknowledged. The work reported in this Account would not have been possible without the excellent efforts by various Ph.D. students, especially Andreas Ohff, Sigmar Pulst, Claudia Lefeber, Normen Peulecke, Dominique Thomas, Frank G. Kirchbauer, Thorsten Zippel, and Paul-Michael Pellny, postdoctoral scientists, technical staff, and many other colleagues whose names appear in the references.

(2)

(3)

(4)

(5)

(6)

References (1) (a) Negishi, E.; Takahashi, T. Organozirconium Compounds as New Reagents and Intermediates. Aldrichim. Acta 1985, 18, 3147. (b) Nugent, W. A.; Thorn, D. L.; Harlow, R. L. Cyclization of Diacetylenes to E,E Exocyclic Dienes. Complementary Procedures 128 ACCOUNTS OF CHEMICAL RESEARCH / VOL. 33, NO. 2, 2000

(7)

Based on Titanium and Zirconium Reagents. J. Am. Chem. Soc. 1987, 109, 2788-2796. (c) Negishi, E. Controlled Carbometallation as a New Tool for Carbon-Carbon Bond Formation and Its Application to Cyclization. Acc. Chem. Res. 1987, 20, 65-72. (d) Buchwald, S. L.; Nielsen, R. B. Group 4 Metal Complexes of Benzynes, Cycloalkynes, Acyclic Alkynes, and Alkenes. Chem. Rev. 1988, 88, 1047-1058. (e) Negishi, E.; Takahashi, T. Organozirconium Compounds in Organic Synthesis. Synthesis 1988, 1-19. (f) Buchwald, S. L.; Fisher, R. A. Zirconocene Complexes of Unsaturated Organic Molecules. Chim. Scr. 1989, 29, 417-421. (g) Negishi, E. Cyclization, Coupling, and Rearrangement Reactions of Organozirconium and Related Compounds. Chim. Scr. 1989, 29, 457-468. (h) Erker, G. Metallocene-Carbene Complexes and Related Compounds of Titanium, Zirconium, and Hafnium. Angew. Chem., Int. Ed. Engl. 1989, 28, 397-406. (i) Buchwald, S. L.; Broene, R. D. Zirconocene Complexes of Unsaturated Organic Molecules: New Vehicles for Organic Synthesis. Science 1993, 1696-1701. (j) Negishi, E.; Takahashi, T. Pattern of Stoichiometric and Catalytic Reactions of Organozirconium and Related Complexes of Synthetic Interest. Acc. Chem. Res. 1994, 27, 124-130. (k) Hanzawa, Y.; Ito, H.; Taguchi, T. Formation of Carbon-Carbon Bonds Using Zirconocene-Butene Complex (“Cp2Zr”) as a Synthetic Tool. Synlett 1995, 299-305. (l) Maier, E. L. C-C-Verknu¨ pfungen mit Metallocenen der vierten Nebengruppe. Nachr. Chem. Tech. Lab. 1993, 41, 811-823. (m) Negishi, E.; Takahashi, T. Alkene and Alkyne Complexes of Zirconocene. Their Preparation, Structure, and Novel Transformations. Bull. Chem. Soc. Jpn. 1998, 71, 755-769. (n) Beckhaus, R. Reaktivita¨ tsprinzipien von Metallocenderivaten fru¨ her U ¨ bergangsmetalle. Nachr. Chem. Tech. Lab. 1998, 46, 611-623. (o) Negishi, E.; Montchamp, J.-L.: Zirconocenes. In Metallocenes, Synthesis, Reactivity, Applications; Togni, A., Halterman, R. L., Eds.; Wiley-VCH: Weinheim, 1998; Vol. 1, pp 241-312. (p) Beckhaus, R.: Titanocenes. In Metallocenes, Synthesis, Reactivity, Applications; Togni, A., Halterman, R. L., Eds.; Wiley-VCH: Weinheim, 1998; Vol. 1, pp 153230. (a) Ohff, A.; Pulst, S.; Peulecke, N.; Arndt, P.; Burlakov, V. V.; Rosenthal, U. Unusual Reactions of Titanocene- und ZirconoceneGenerating Complexes. Synlett 1996, 111-118. (b) Rosenthal, U.; Ohff, A.; Michalik, M.; Go¨ rls, H.; Burlakov, V. V.; Shur, V. B. Transformation of the First Alkyne Complex without an Additional Phosphane Ligand into a Dinuclear σ-Alkenyl Complex by Hydrogen Transfer from η5-C5H5 to the Alkyne Ligand. Angew. Chem., Int. Ed. Engl. 1993, 32, 1193-1195. (c) Burlakov, V. V.; Polyakov, A. V.; Yanovsky, A. I.; Struchkov, Y. T.; Shur, V. B.; Vol’pin, M. E.; Rosenthal, U.; Go¨ rls, H. Novel Acetylene Complexes of Titanocene and Permethyltitanocene without Additional Ligands. Synthesis, Spectral Characteristics and X-ray Diffraction Study. J. Organomet. Chem. 1994, 476, 197-206. (d) Mansel, S.; Thomas, D.; Lefeber, C.; Heller, D.; Kempe, R.; Baumann, W.; Rosenthal, U. Stable rac-[1,2-Ethylene-1,1′-bis(η5-tetrahydroindenyl)]-zirconacyclopentanes. Synthesis, X-ray Crystal Structures, Stability, and Catalysis of Ethylene Polymerization. Organometallics 1997, 16, 2886-2890. (a) Hitchcock, P. B.; Kerton, F. M.; Lawless, G. A. The Elusive Titanocene. J. Am. Chem. Soc. 1998, 120, 10264-10265. (b) Horacek, M.; Kupfer, V.; Thewalt, U.; Stepnicka, P.; Polasek, M.; Mach, K. Bis[η5-tetramethyl(trimethylsilyl)-cyclopentadienyl]titanium(II) and Its π-Complexes with Bis(trimethylsilyl)acetylene and Ethylene. Organometallics 1999, 18, 3572-3578. Diederich, F.: Oligoacetylenes. In Modern Acetylene Chemistry; Stang, P. J., Diederich, F., Eds.; VCH: Weinheim, 1995; pp 443469. (a) Mohamadi, F.; Spees, M. Stable Bis(η5-cyclopentadienyl)zirconacyclopentadiene Complexes. Organometallics 1992, 11, 1398-1400. (b) Mathieu, A.; Miguel, Y.; Igau, A.; Donnadieu, B.; Majoral, J.-P. Versatile Behaviour of a New Class of Bicyclic Compounds: Zirconacyclopentadiene Phosphiranes. Organometallics 1997, 16, 3086-3088. (c) Metzler, N.; No¨ th, H.; Thomann, M. Zirconocene-Mediated Cyclization of Alkynyl-Diborylmethanes to 1,3-Diborolanes. Organometallics 1993, 12, 2423-2425. (a) Mao, S. S. H.; Liu, F.-Q.; Tilley, D. T. Efficient ZirconoceneCoupling of Silicon-Substituted Diynes to Polymers and Macrocycles. J. Am. Chem. Soc. 1998, 120, 1193-1206. (b) Lucht, B. L.; Mao, S. S. H.; Tilley, D. T. A Zirconocene-Coupling Route to Substituted Poly(p-phenylene)s: Band Gap Tuning via Conformational Control. Bipyridine-Containing Cyclophanes via Zirconocene Coupling. J. Am. Chem. Soc. 1998, 120, 4354-4365. (c) Nitschke, J.; Tilley, D. T. Bipyridine-Containing Cyclophanes via Zirconocene Coupling J. Org. Chem. 1998, 63, 3673-3676. Du, B.; Farona, M. F.; McConnville, D. B.; Youngs, W. J. Synthesis and Structure of a Tricyclic Bis(zirconacyclopentadiene) Compound. Tetrahedron 1995, 51, 4359-4370.

Titano- and Zirconocenes with Diynes and Polyynes Rosenthal et al. (8) Xi, Z.; Fischer, R.; Hara, R.; Sun, W.-H.; Obora, Y.; Suzuki, N.; Nakajima, K.; Takahashi, T. Zirconocene-Mediated Intramolecular Carbon-Carbon Bond Formation of Two Alkynyl Groups of Bis(alkynyl)silanes J. Am. Chem. Soc. 1997, 119, 12842-12848. (9) Warner, B. P.; Davis, W. M.; Buchwald, S. L. Synthesis and X-ray Structure of a Zirconocene Complex of Two Alkynes. J. Am. Chem. Soc. 1994, 116, 5471-5472. (10) Tillack, A.; Baumann, W.; Ohff, A.; Lefeber, C.; Spannenberg, A.; Rosenthal, U. Intramolecular Cyclization of Terminal Disubstituted R,ω-Diynes with Titanocene “Cp2Ti” Giving a Subsequent Unusual Cp-Ring-Opening and a New Intramolecular C-C Coupling Reaction. J. Organomet. Chem. 1996, 520, 187-193. (11) Tomaszewski, R.; Hyla-Kryspin, I.; Mayne, C.; Arif, A. M.; Gleiter, R.; Ernst, R. D. Shorter Nonbonded than Bonded Contacts or Nonclassical Metal-to-Carbon Atom Interactions? J. Am. Chem. Soc. 1998, 120, 2959-2960. (12) Liu, F.-Q.; Harder, G.; Tilley, D. T. A Simple, Efficient Route to Cage Compounds via Zirconocene Coupling. J. Am. Chem. Soc. 1998, 120, 3271-3272. (13) Hsu, D. P.; Davis, W. M.; Buchwald, S. L. Synthesis and Structure of a Seven-Membered Cyclic Cumulene. J. Am. Chem. Soc. 1993, 115, 10394-10395. (14) Burlakov, V. V.; Ohff, A.; Lefeber, C.; Tillack, A.; Baumann, W.; Kempe, R.; Rosenthal, U. The First Titanacyclic Five-Membered Cumulene. Synthesis, Structure, and Reactivity. Chem. Ber. 1995, 128, 967-971. (15) (a) Rosenthal, U.; Go¨ rls, H. The Formation of [Cp2Ti(CtCSiMe3)]2 from Titanocene and 1,4-Bis(trimethylsilyl)-1,3-butadiyne. J. Organomet. Chem. 1992, 426, C53-C59. (b) Rosenthal, U.; Ohff, A.; Baumann, W.; Kempe, R.; Tillack, A.; Burlakov, V. V. Reaction of Disubstituted 1,3-Butadiynes R1CtCCtCR2 with Zirconocene Complexes: Cleavage of the Central CsC Single Bond To Form Symmetrically and Unsymmetrically Doubly Acetylide-Bridged Metallocene Complexes. Organometallics 1994, 13, 2903-2906. (16) Pavankumar, P. N. V.; Jemmis, E. D. Structure and Bonding in L2M(µ-CCR)ML2 and L2M(µ-RC4R)ML2 Complexes (L2M ) Cp2Ti, Cp2Zr, R2Al, R(NR3)Be, (tmpda)Li, H2B, and H2C+). A Molecular Orbital Study. J. Am. Chem. Soc. 1988, 110, 125. (17) (a) Teuben, J. H.; De Liefde Meijer, H. J. Phenylethinyl Complexes of Titanocene and Vanadocene. J. Organomet. Chem. 1969, 17, 87. (b) Sekutowsky, D. G.; Stucky, G. D. Oxidative Coupling of the Phenylethynyl Group in µ-(1-3η:2-4η-trans,trans-1,4-Diphenyl1,3-butadiene with Bis(cyclopentadienyl)titanium(II). J. Am. Chem. Soc. 1976, 98, 1376. (c) Rosenthal, U.; Ohff, A.; Tillack, A.; Baumann, W.; Go¨ rls, H. Zu Reaktionen disubstituierter Butadiine mit Titanocen-Komplexen. J. Organomet. Chem. 1994, 468, C4C8. (18) Jemmis, E. D.; Giju, K. T. To Couple or Not To Couple: The Dilemma of Acetylide Carbons in [(η5-C5H5)2M(µ-CCR)M(η5-C5H5)2] Complexes (M ) Ti, Zr)sA Theoretical Study for R ) H, F. Angew. Chem., Int. Ed. Engl. 1997, 36, 606-608. (19) Burlakov, V. V.; Peulecke, N.; Baumann, W.; Spannenberg, A.; Kempe, R.; Rosenthal, U. Reactions of the Phenyl Substituted FiveMembered TitanacyclocumulenesUnusual Coupling of a 1,4Disubstituted 1,3-Butadiyne with Two Titanium atoms. J. Organomet. Chem. 1997, 536-537, 293-297. (20) Rosenthal, U.; Ohff, A.; Baumann, W.; Kempe, R.; Tillack, A.; Burlakov, V. V. Synthesis and Structure of the Smallest Cyclic Cumulene; Reaction of 1,3-Diynes with Zirconocene Complexes. Angew. Chem., Int. Ed. Engl. 1994, 33, 1605-1607. (21) (a) Johnson, R. P. Strained Cyclic Cumulenes. Chem. Rev. 1989, 89, 1111-1124. (b) Angus, R. O., Jr.; Johnson, R. P. Small-Ring Cyclic Cumulenes: The Structure and Energetics of Cyclic Butatrienes and the Synthesis of 1,2,3-Cyclononatriene. J. Org. Chem. 1984, 49, 2880-2883. (22) Jemmis, E. D.; Giju, K. T. Novel Mechanism for Interesting C-C Coupling and Cleavage Reactions and Control of Thermodynamic Stability Involving [L2M(µ-CCR)ML2] and [L2M(µ-RCC-CCR)ML2] Complexes (M ) Ti, Zr; L ) η5-C5H5, Cl, H.; R ) H, F, CN): A Theoretical Study. J. Am. Chem. Soc. 1998, 120, 6952-6964. (23) Pellny, P.-M.; Burlakov, V. V.; Peulecke, N.; Baumann, W.; Spannenberg, A.; Kempe, R.; Francke, V.; Rosenthal, U. Dimerization of Titanacyclocumulenes to Titanium Substituted Radialenes.

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

Synthesis, Stability, and Reactions of Five-Membered Titanacyclocumulenes with a Coupling of Two 1,4-Diphenyl-1,3-butadiyne between Two Titanocene Molecules to Radialene-Like Fused Titanacyclopentadiene Compounds. J. Organomet. Chem. 1999, 578, 125-132. Pulst, S.; Arndt, P.; Heller, B.; Baumann, W.; Kempe, R.; Rosenthal, U. Nickel(0) Complexes of Five-Membered Titana- and Zirconacyclocumulenes as Intermediates in the Cleavage of C-C Bonds of Disubstituted Butadiynes. Angew. Chem., Int. Ed. Engl. 1996, 35, 1112. Rosenthal, U.; Pulst, S.; Arndt, P.; Ohff, A.; Tillack, A.; Baumann, W.; Kempe, R.; Burlakov, V. V. Heterobimetallic σ,π-AcetylideBridged Complexes from Disubstituted 1,3-Butadiynes. Organometallics 1995, 14, 2961-2968. (a) Jones, M.; Klosin, J. Transition Metal Complexes of Arynes, Strained Cyclic Alkynes, and Strained Cyclic Cumulenes. Adv. Organomet. Chem. 1998, 42, 147. (b) Danjoy, C.; Zhao, J.; Donnadieu, B, Legros, J.-P.; Valade, L.; Choukroun, R.; Zwick, A.; Cassoux, P. [Cp2V(µ-η2: η4-butadiyne)MCp′2](M ) Ti, Zr; Cp′ ) C5H5, C5H4SiMe3): Heterodimetallic Complexes from a Bis(alkynyl)metallocene and VanadocenesUnexpected Structure of [Cp2V(µ-η2:η4-butadiyne)Zr(C5H4SiMe3)2] Containing Two Planar-Tetracoordinate Carbon Atoms. Chem. Eur. J. 1998, 4, 1100-1105. (c) Choukroun, R.; Cassoux, P. Planar Tetracoordination of Carbon in Groups 4 and 5 Organometallic Chemistry. Acc. Chem. Res. 1999, 32, 494-502. (a) Fro¨ mberg, W. Ph.D. Thesis, University of Bochum, 1986. (b) Temme, B.; Erker, G.; Fro¨ hlich, R.; Grehl, M. Heterobimetallic Betaine Chemistry: Catalytic and Stoichiometric Couplings of Alkynyl Ligands under the Influence of Both Zirconium and Bor Centers. Angew. Chem., Int. Ed. Engl. 1994, 33, 1480-1482. (c) Erker, G.; Fro¨ mberg, W.; Benn, R.; Mynott, R.; Angermund, K.; Kru¨ ger, C. p-Conjugation and Dynamic Behavior in Doubly Acetylide-Bridged Binuclear Group 4 Bent Metallocene Complexes. Organometallics 1989, 8, 911-920. Kirchbauer, F. G.; Pulst, S.; Heller, B.; Baumann, W.; Rosenthal, U. The first Metathesis of C-C Single Bonds in Homogeneous Solution: Titanocene Mediated and Photocatalyzed Cleavage and Recombination of Disubstituted Butadiynes. Angew. Chem., Int. Ed. 1998, 37, 1925-1927. Vidal, V.; Theolier, A.; Thivolle-Cazat, J.; Basset, J.-M. Metathesis of Alkanes Catalyzed by Silica-Supported Transition Metal Hydrides. Science 1997, 276, 99. (a) Pellny, P.-M.; Kirchbauer, F. G.; Burlakov, V. V.; Baumann, W.; Spannenberg, A.; Rosenthal, U.: Reactivity of Permethylzirconocene and -titanocene towards 1,3-Butadiynes: η4- vs η2 Complexation or C-C Coupling with the Permethyltitanocene. J. Am. Chem. Soc. 1999, 121, 8313-8323. (b) Horacek, M.; Stenicka, P.; Gyepes, R.; Cisarova, I.; Polasek, M.; Mach, K.; Pellny, P.-M.; Burlakov, V. V.; Baumann, W.; Spannenberg, A.; Rosenthal, U.: Novel Addition Reactions of 2,2,7,7-Tetramethyl-3,5-octadiyne to the Methyl Groups of a η5-Pentamethylcyclopentadiynyl Ligand. J. Am. Chem. Soc. 1999, 121, 10638-10639. Pellny, P.-M.; Kirchbauer, F. G.; Burlakov, V. V.; Baumann, W.; Spannenberg, A.; Rosenthal, U.: Different C-C Coupling Reactions of Permethylzirconocene and -titanocene with Disubstituted 1,3-Butadiynes. Chem. Eur. J. 2000, 6, 81-90. Pellny, P.-M.; Peulecke, N.; Burlakov, V. V.; Tillack, A.; Baumann, W.; Spannenberg, A.; Kempe, R.; Rosenthal, U. Twofold C-C Single Bond Activation and Cleavage in the Reaction of Octatetraynes with Titanocene and Zirconocene Complexes. Angew. Chem. Int. Ed. Engl. 1997, 36, 2615-2617. Pellny, P.-M.; Burlakov, V. V.; Baumann, W.; Spannenberg, A.; Kempe, R.; Rosenthal, U. Novel Ti6-, Zr3-, and Zr6-Complexes from Branched Polyynes and Titanocene as well as Zirconocene. Organometallics 1999, 18, 2906-2909.

AR9900109


What Do Titano- and Zirconocenes Do with Diynes and Polyynes

Recommend Documents